SMOOTH / ROUGH

An interactive study of how smooth and rough move from physical texture into metaphorical language

Click the black side tabs to turn pages.

*Source data: Gutiérrez et al. (2016) annotated 8,592 adjective–noun phrases as literal or metaphorical, using combinations occurring at least ten times in a 4.58-billion-token corpus. This site isolates the smooth and rough subset. Each noun is a recorded partner in a phrase such as smooth operation or rough estimate; frequency is its corpus count.

*On the map, scroll or drag vertically to reveal more nouns. Use the filters to compare shared nouns, concepts, and themes.

*In the chart, hover or tap a mark for details. Click to pin a card, drag its title to move it, and use × to close it.

Smooth (n.)

Shared (n.)

Rough (n.)

SMOOTH AS REMOVAL ROUGH AS MARKING

At the literal level, smooth and rough operate as expected opposites across surfaces, materials, and physical contact.

In metaphorical uses, the opposition becomes asymmetric. Smooth tends to describe friction, interruption, or irregularity after it has been reduced: a smooth operation, transition, or performance. Rough more often makes a residual condition explicit: a rough passage, time, draft, or estimate.

Shared nouns show that the contrast belongs to context rather than to the noun alone. Their horizontal position records the balance between smooth and rough frequency; size records their combined frequency.

Smooth presents the condition after friction is reduced. Rough marks what remains unresolved.

This is a pattern in this dataset, not a universal definition.

REFERENCES

Dataset: Gutiérrez et al. (2016), AN phrase annotations. This website uses the smooth and rough subset of the authors’ annotated adjective–noun data.

Processing note: AI-assisted methods were used to help organize the large set of phrases into provisional semantic groups and to structure the interactive data. The groupings and representative examples were manually reviewed and revised; the source dataset’s literal/metaphorical labels and frequency counts were retained.

[1]

Gutiérrez, E. D., Shutova, E., Marghetis, T., & Bergen, B. K. (2016). Literal and metaphorical senses in compositional distributional semantic models. In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) (pp. 183–193). Association for Computational Linguistics. https://doi.org/10.18653/v1/P16-1018

[2]

Kashkin, E., & Vinogradova, O. (2015). The domain of surface texture. Basic Research Program Working Papers, 30/LNG/2015. National Research University Higher School of Economics. https://doi.org/10.2139/ssrn.2698127

[3]

Hurtienne, J., & Reinhardt, D. (2017). Texture metaphors and tangible interaction: No smooth relationship? In Proceedings of the Eleventh International Conference on Tangible, Embedded, and Embodied Interaction (pp. 79–87). Association for Computing Machinery. https://doi.org/10.1145/3024969.3024986

[4]

Nagano, H., Okamoto, S., & Yamada, Y. (2014). Semantically layered structure of tactile textures. In M. Auvray & C. Duriez (Eds.), Haptics: Neuroscience, devices, modeling, and applications (pp. 3–9). Springer. https://doi.org/10.1007/978-3-662-44193-0_1